Parallel data in the domain of interest is the key resource when training a statistical machine translation (SMT) system for a specific purpose. Since ad-hoc manual translation c...
Prasanth Kolachina, Nicola Cancedda, Marc Dymetman...
We examine the relationship between the VCdimension and the number of parameters of a smoothly parametrized function class. We show that the VC-dimension of such a function class ...
Wee Sun Lee, Peter L. Bartlett, Robert C. Williams...
We report on an extension of Haskell with open type-level functions and equality constraints that unifies earlier work on GADTs, functional dependencies, and associated types. The...
Tom Schrijvers, Simon L. Peyton Jones, Manuel M. T...
Overfitting is a fundamental problem of most machine learning techniques, including genetic programming (GP). Canary functions have been introduced in the literature as a concept ...
We propose a new method for comparing learning algorithms on multiple tasks which is based on a novel non-parametric test that we call the Poisson binomial test. The key aspect of...